NASA Technical Memorandum !O3180 ..........

_ _ =_ Sequentially Evaporated Thin Film ..........__..............YBa2Cu307-x Superconducting

Microwave Ring Resonator

Norman J. Rohrer, Hing Y.To, and George J. Valco

Ohio State University .....

Columbus, Ohio

Kul B. Bhasin ..................

Lewis Research Center

Cleveland, Ohio

Chris Chorey

Sverdrup Technology, Inc.Lewis Research Center Group

Brook Park, Ohio

Joseph D. WarnerLewis Research Center

Cleveland, Ohio

Prepared for the _Conference on the Science and Technology of Thin Film Superconductors--

, sponsored by the U.S. Department of EnergyDenver, Colorado, April 30--May 4, 1990

(NASA-TM-I031gO) SFQUENTIALLY EVAPORA/fD

THIN FILM y_2Cu3_(7-x) SUPERCONDUCTING

MICPOWAVE RING RESONATOR (NASA) 12 pCSCL 09A

N90-25273

G31_3

Unclas

02B_154

https://ntrs.nasa.gov/search.jsp?R=19900015957 2020-06-04T17:15:13+00:00Z

SEQUENTIALLY EVAPORATED THIN FILM YBa2Cu307_ x

SUPERCONDUCTING MICROWAVE RING RESONATOR

Norman J. Rohrer, King Y. To, and George J. Valco

Department of Electrical Engineering

Ohio State University, Columbus, Ohio 43210

Kul B. Bhasin

National Aeronautics and Space Administration

Lewis Research Center, Cleveland, Ohio 44135

Chris Chorey

Sverdrup Technology, Inc.

Lewis Research Center Group, Brook Park, Ohio 44142

Joseph D. Warner

National Aeronautics and Space Administration

Lewis Research Center, Cleveland, Ohio 44135

ABSTRACT

There is great interest in the application of thin film

high temperature superconductors in high frequency electronic

circuits. A ring resonator provides a good test vehicle for

assessing the microwave losses in the superconductor and for

comparing films made by different techniques. Ring resonators

made of YBa2Cu3OT_ x have been investigated on LaAIO 3 substrates.

The superconducting thin films were deposited by sequential

electron beam evaporation of Cu, Y, and BaF2 with a post

anneal. Patterning of the superconducting film was done using

negative photolithography. A ring resonator was also

fabricated from a thin gold film as a control. Both resonators

had a gold ground plane on the backside of the substrate. Thering resonators' reflection coefficients were measured as a

function of frequency from 33 to 37 GHz at temperatures rangingfrom 20 K to 68 K. The resonator exhibited two resonances

which were at 34.5 and 35.7 GHz at 68 K. The resonant

frequencies increased with decreasing temperature. Themagnitude of the reflection coefficients is used in the

calculation of the unloaded Q-values. The performance of the

evaporated and gold resonator are compared with the performance

of a laser ablated YBa2Cu307-x resonator. The causes of the

double resonance are discussed.

INTRODUCTION

The advent of high temperature superconductors has drawn

attention towards the possibilities of using thin films

superconductors in microwave circuits. Several measurement

techniques have been employed for characterization of the films

including high Q cavities [I], stripline resonators [2] and

ring resonators [3]. The surface resistance of the the

YBa2Cu307-x superconductor has been investigated on bulk

samples, thin films and single crystals [4-6]. In this paper

we employ a ring resonator to study the microwave properties of

YBa2Cu307_ 8 fabricated by multi-layer sequential evaporation

with post-anneal.

DEPOSITION AND ANNEAL PROCEDURES

Electron beam evaporation was used for deposition of Cu,

Y, and BaF 2 on LaAIO 3. The materials were deposited in that

order which was repeated four times for a total of twelve

layers. The thicknesses of the individual layers were 507

angstroms for Cu, 473 angstroms for Y, and 1704 angstroms for

BaF2. The details of the deposition process have been

previously reported [7,8].

The multilayer stack was subjected to a post anneal to

assist in the formation of the proper phase of YBa2Cu3OT-x. The

samples were inserted into a preheated furnace using a slow

push. They were annealed at 900 oC for 45 minutes in oxygen

bubbled through room temperature water. The temperature was

then ramped down to 450 °C where it was held for six hours.

Finally, the temperature was ramped down to room temperature.

The ambient was dry oxygen during all stages except the high

temperature anneal. This procedure resulted in a one micron

thick superconducting thin film with an onset temperature of 93

K and a critical temperature of 85 K.

PATTERNING PROCEDURE

The ring resonator was patterned using negative

photolithography. KTI 752 photoresist was spun on at a rate of

4,000 rpm for 60 Seconds which resulted the photoresist being

1.7 _m thick. The sample was soft baked at 95 °C for 25

minutes and exposed through a dark field mask for five seconds

with an illumination power density of 34.8 mW/cm 2. The

photoresist was developed for 2 minutes and 45 seconds in 802

developer and rinsed in ethanol. A one percent molar bromine

solution in ethanol was used for etching the thin film followed

by a 30 second rinse. Finally, the photoresist was removed in

SN-10 stripper.

Once the superconductor was patterned,-a ground plane was

deposited on the back of the substrate using electron beam

evaporation. For adhesion, a 1400 angstrom titanium layer was

deposited before the one micron gold ground plane.

RESONATOR ANALYSIS

The following dimensions correspond to the resonator shown

in Figure I. H is the substrate thickness.

H = 254 microns

S - 36 microns

W - 143.3 microns

R - (rl + r2)/2 - 990 microns

s I

Figure 1: Ring resonator.

The resonator was designed for 50 ohm transmission lines

with a resonant frequency of 30 GHz for the first harmonic. At

the time of the design, the dielectric constant for LaAIO3 was

reported to be 15. Since then a more recent value reported forthe dielectric constant is 21.9. This has resulted in a

characteristic impedance for this geometry 42.9 ohms. The

first harmonic was decreased below K-band. For this reason, we

measured at the second harmonic which had a resonant frequency

of 35.1 GHz at 25 K.

Microwave Testing

Once fabricated the ring resonator was experimentally

tested using an HP 8510B network analyzer. The microwave test

setup was configured using waveguides. Thus, a waveguide to

microstrip transition was implemented using a cosine tapered

ridge transition [9].

0

.u_i -s

u -I0

"_u -15

_ -20

33 34 35 36 37

Frequency (GHz)

Figure 2: Magnitude of the reflection coefficient

for ring resonator's fabricated from gold and

YBa2Cu3OT- x at 30 K.

3

The reflection coefficient for the resonators was measured as a

function of frequency at temperatures ranging from 20 K to 300

K using a CTI closed cycle cryogenic refrigerator. The

superconducting resonator was measured from 20 K to its

critical temperature. The resonant frequency superconducting

resonator was 35.1 GHz at 25 K. Figure 2 shows the magnitude

of the reflection coefficient for both the gold and the

superconducting resonator at approximately 30 K.

The superconducting resonator exhibited two resonant

valleys. Valley 1 matched the resonant frequency of the gold

resonator while valley 2 occurred at 36.1 GHz. As will be

discussed below, the occurrence of the second valley can be

caused by one section of the ring resonator having a larger

impedance than the rest of the ring.

Double Resonance Modeling

A transmission line model was implemented in Touchstone

[10] to allow study of the double resonance. The resonator was

simulated using transmission lines to match the physical layout

and capacitors to model the gap. The capacitance values used

were calculated from empirical equations derived in reference

[ii]. The impedance was increased in a region corresponding to

three percent of the ring's circumference located closer to the

transmission end. The location of this section for our model

was 114 ° from the input. The impedance of this region was

increased from a single line with 42.9 ohms to two parallel

lines of 150 ohms each. This simulates a blister centered in

the ring's transmission line. Figure 3 presents the resultsfrom this model.

0

o

U _ -I0

O-_

o

----o---- Touchstone Model

-20 ------_ I L.._ ! , 1 , I ,

34.5 35.0 35.5 36.0 36.5 37.0

Frequency (GHz)

Figure 3: Modeled ring resonator with blister

centered in the transmission line. Impedance of the

lines around the blister were 150 ohms. Blister was

located 1/3 the way around thering.

The resonant frequencies of the model match directly with

the resonant frequencies of the superconducting ring resonator.

The width of valley 1 is quite similar, but the width of valley

2 is smaller for the modeled resonator. The reflection

coefficient is much lower off resonance for the superconducting

resonator. The the depth of the valley's peaks could be

altered by varying the position of the blister around the ring.

The position used for the calculation shown in Figure 3 agrees

with the location of a blistered region observed in the film of

the resonator. The separation of the two valleys could be

increased by increasing the impedance of the transmission lines

around the blister. The response of the model remained thesame for different number of transmission lines used within the

high impedance region as long as the parallel combination of

the characteristic impedances remained constant.

Resonant Freuuency Shifts

The group velocity for a microstrip configuration with

superconducting transmission lines varies as a function of

temperature for temperatures less than the superconductor's

critical temperature. This corresponds directly to a change in

wavelength. Thus, the resonant frequency also varies as a

function of temperature. The group velocity for a

superconducting transmission line with a superconducting ground

plane is given by [12]

V m

where II and tl are the transmission line's penetration depth

and thickness, respectively, and k2 and t2 are the ground

plane's penetration depth and thickness, respectively. If the

ground plane is a normal metal, the group velocity is reducedto

V _--"

c

_/eeff(f)+ cot

Note that the first equation is general enough to accommodate

microstrip circuitry with superconductors of different

penetration depths for the transmission line and ground plane.

Figure 4 shows the calculated resonant frequency for a 0.7

micron thick film as a function of the temperature normalized

to the critical temperature. This graph shows three plots.

Two plots show the comparison of the resonant frequency with

two different penetration depths for a sample with a gold

ground plane and a superconducting resonator. The penetration

depths were chosen to be on each side of the values

experimentally determined by [13]. The third plot represented

by the open squares shows the resonant frequency as a function

of temperature for a sample with both the ground plane and the

resonator being superconducting. Replacing both the

transmission lines and the ground plane with a superconductor

will decrease the losses in the circuit if the superconductor

losses are lower than the gold losses. The resonant frequency

for a sample with both a superconducting transmission llne and

a ground plane will exhibit a larger shift in the resonant

frequency as the temperature nears the critical temperature for

the superconducting film.

5

N

v

36.5 I

35.5

34.5

33.5

0.75

/Both Ground Plane and

0.80 0.85 0.90 0.95 1.00

T/Tc

"-¢--- 2500 Angstroms

-'-I---5000 Angstroms

---9-- 5000 Angstroms

Figure 4: Resonant frequency versus normalized

temperature for a 0.7 micron thick film. Different

penetration depths are shown as well as an example of

both ground plane and transmission lines beingsuperconductors.

The shift in the resonant frequency for the sequentially

evaporated superconducting ring resonator exhibited the same

shape as the theoretical predictions. The measured resonant

frequency as a function of temperature is given in Figure 5.

36.5

N

v

36.0

>.u 35.5

35.0

34.5

• • •

m [][] m

[][]

! l " i I'a "

20 40 60 80

Temperature (K)

m Valley 1

• Valley 2

Figure 5: Resonant frequency versus temperature for

the superconducting ring resonator. Both valleys are

represented.

An attempt was performed to try to match the theoretical

equations with the experimental data. The equations resulted

in a penetration depth that was much larger than the thickness

of the film. This is not reasonable since the film would no

longer be superconducting for a large penetration depth. A

possible explanation for the large penetration depth is due to

the film being granular. This may allow more penetration at

between grains.

Unloaded 0 Calculation

The unloaded Q of the resonator can be extracted from the

reflection coefficient. The model used for the derivation of

the equation is a parallel RLC circuit with an ideal

transformer in parallel and a series input resistance [14].This model is for an unterminated resonator. It was used since

the resonator was not loaded on the transmission side during

our testing.

The derivation of the

published in reference [14].

given by

QL =

equations presented here are

The loaded Q of a resonator is

fr

fl - f2

where fl and f2 are the half power points on each side of the

resonant frequency ft. The values for the half power levels

are calculated by

1 , - (;- 1

La+ ill

where _ is the coupling loss and k' is the effective coupling

coefficient. The coupling loss is calculated far off resonance

where the reflection coefficient (Fi) is nearly constant.

1 - F i_=

1 +F i

The effective coupling coefficient is the sum of the coupling

coefficient and the coupling loss. The coupling coefficient

can be easily calculated using the reflection coefficient (Fr)

at the resonant frequency. The coupling coefficient can be

calculated by

1 - F r 1 + F rk =. or k-

1 + F r i - F r

for the undercoupled and overcoupled cases, respectively.

The unloaded Q can be calculated from the loaded Q by

Qo = QL(l]++k'.)(y

The unloaded Q as a function of temperature is shown in

Figure 6. The unloaded Q for both valleys of the sequentially

evaporated resonator are shown.

O

4O0

300

200

I00

m •[] []

[]

D

D []D

[]0

UU•]mum •

m

I " " " " I " • " " I

O 25 50 75 100

Temperature (K)

• Gold

o Valley 1

• Valley 2

Figure 6: Unloaded Q versus temperature for the gold

ring and the superconducting ring. Both valleys are

shown for the superconducting ring.

The gold resonator had a larger unloaded Q than our

sequentially evaporated film. At iower temperatures the

difference between the Unloaded Q values narrowed since the

superconducting resonator's unloaded Q increased faster then

the gold resonator's when compared to valley I.

Surface Resistance

The surface resistance of the ring resonator can be

extracted from the unloaded Q. The surface resistance is givenby [15]

4KZo _ ( 1 1 )RSS-- RSAU B(C + D) _, QJAU Q%S

where RSA u is the surface resistance of gold, and QoAU and Qos

are the unloaded Q values of the gold resonator and the

superconducting resonator at the same temperature,

respectively. The constants B, C, and D are related to the

physical dimensions and are given in reference [15].

The surface resistance as a function of temperature for

both the gold and the sequentially evaporated films are shown

in Figure 7. For comparison a ring resonator was fabricated

from a film deposited by laser ablation [16]. The surface

resistance calculated for this film is also shown in Figure 7.

The graph shows that the sequentially evaporated film had

the highest surface resistance at all temperatures. The gold

film's surface resistance was about two-thirds the value of the

sequentially evaporated film at 25 K. The laser ablated fiim

had a surface resistance of approximately half that of gold at

temperatures less than 50 K. As the temperature neared the

critical temperature, the surface resistance of the laser

ablated film started to increase rapidly to a value larger than

that for gold at 70 K.

8

0.08

0

¢ 0.(16U

0 04-,-4U)

o_ 0 02

U

u,4

0 O0

_o 0

mm[]

m Dm m _B • BB_ _B

0OO • • •

• I • • " ' i " " i " " ' "

25 50 75 I00

[] Gold

• Laser Ablated

• Sequential Dep.

Temperature (K)

Figure 7: Surface resistance versus temperature for

a gold resonator and two superconducting resonators.

The superconducting resonators were deposited using

different techniques. One was deposited by laser

ablation and one by sequential evaporation.

CONCLUSION

A sequentially evaporated YBa2Cu307_ x superconducting thin

film was patterned into a ring resonator using negative

photolithography. The ring resonators' reflection coefficients

were measured using an HP 8510B network analyzer as a function

of frequency from 33 to 37 GHz at temperatures ranging from 20

K to 68 K. The resonator exhibited two resonances which were

at 34.5 and 35.7 GHz at 68 K. The resonant frequencies

increased with decreasing temperature. The double resonance

could be explained using a model that allowed for a small

section of the ring resonator to have a larger impedance than

the rest of the ring resonator. The location of the high

impedance section in the model correlated well with the

location of a blistered region of the film in the resonator.

Once the reflection coefficient data was taken, the unloaded Q

was extracted. The superconducting resonator was compared to a

gold resonator. The gold resonator had a higher unloaded Q

value at all temperatures• This translated into the gold

having a lower surface resistance than the sequentially

evaporated superconducting film. The surface resistance of the

gold was about two-thirds the surface resistance of the

sequentially evaporated superconducting film at 25 K. When

compared to the laser ablated film, the laser ablated film's

surface resistance was about one half that of gold for

temperatures less than 50 K.

ACKNOWLEDGEMENTS

Design of the ring resonator was done by Robert Romanofsky

and one of the authors (Kul Bhasin). This research was funded

by the National Aeronautics and Space Administration, Lewis

9

Research Center under cooperative research agreement NCC 3-105

and under the graduate student research program NGT 50378.

REFERENCES

I. J. S. Marten, J. B. Beyer, and D. S. Ginley, Appl. Phys.

Lett. 52 (21), 1822 (1988).

2. M. DiIorio, A. C. Anderson, and B. Y. Tsaur, Phys. Rev.

B., 38 (10), 9726 1988.

3. J. H. Takemoto, F. K. Oshita, H. R. Fetterman, P. Korbin,

E. Sovero, MTT-37, 1650 (1989).

4. M. K. Wu, J. R. Ashburn, C. J. Torng, P. H. Hor, R. L.

Meng, L. Gao, Z. J. Huang, Y. O. Wang, and C. W. Chue,

Phys. Rev. Lett._ 58,908 (1987).5. T. M. P. Percival, J. S. Thorn, and R. Driver, Electron.

Lett., 23, 1225 (1987).

6. I. Sankawa, M. Sato, T. Konaka, M. Dobayashi, and K.

Ishihara, Jap. Journ. Appl. Phys., 27 (9), L1637 (1988).

7. G. J. Valco, N. J. Rohrer, J. D. Warner and K. B. Bhasin,

_[ig/l_c Superconducting Thin ?iims. Devices and

Applications, G. Margaritiondo, R. Joynt and M. Onellion,

AIP Conference Proceedings No. 182, pp. 147-154, American

Institute of Physics, Atlanta, 1988.

8. G. J. Valco, N. J. Rohrer, J. D. Warner and K. B. Bhasin,

_.__II__D__/_hg__op on High Temperature

Superconductivity, pp. 197-203, GACIAC, Hunstville, 1989.

9. R. R. Romanofsky and K. A. Shalkhauser, NASA Technical

Paper # 2875, 1989.I0_ Comm@rcia!ly available software, EEsof, West _ _e Village,

CA 91632.

Ii. R. Garg and J. J. Bahl, Int. Journ. Elect., 45 (1978).

12. J. C. Swihart, Journ. Appl. Phys., 32 (3), 451 (1961).

13. S. M. Anlage, H. Sze, H. J. Snortland, S. Tahara, B.

Langley, C. B. Eom, M. R. Beasley, R. Taber, Appl. Phys.

Lett. 54 (26) 1989.

14. Robert R. Romanofsky, NASA Technical Paper #2899, 1989.

15. J. Takamoto, F. Oshita, H. Fetterman, IEEE Trans.,

MTT-37, 1650 (1989).16. K. B. Bhasin, C. M. Chorey, J. D. Warner, R. R.

Romanofsky, V. O. Heinen, K. S. Kong, H. Y. Lee, and T.

Itoh, To be published in Symp. Digest of the 1990 IEEE

Int. Microwave Symp., Dallas, TX.

i0

Report Documentation PageNational Aeronautics andSpace Administration

1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.

NASA TM-103180

5. Report Date4. Title and Subtitle

Sequentially Evaporated Thin Film YBa2Cu3OT_ x Superconducting

Microwave Ring Resonator

7. Author(s)

Norman J. Rohrer, Hing Y. To, George J. Valco, Kul B. Bhasin,

Chris Chorey, and Joseph D. Warner

9. Performing Organization Name and Address

National Aeronautics and Space AdministrationLewis Research Center

Cleveland, Ohio 44135-3191

12. Sponsoring Agency Name and Address

National Aeronautics and Space Administration

Washington, D.C. 20546-0001

6. Performing Organization Code

8. Performing Organization Report No.

E-5557

10. Work Unit No.

UPI307-51-00

11. Contract or Grant No.

13. Type of Reportand PeriodCovered

Technical Memorandum

14. Sponsoring Agency Code

15. Supplementary Notes

Prepared for the Conference on the Science and Technology of Thin Film Superconductors sponsored by the

U.S. Department of Energy, Denver, Colorado, April 30--May 4, 1990. Norman J. Rohrer, Hing Y. To,

George J. Valco, Department of Electrical Engineering, Ohio State University, Columbus, Ohio 43210.

Kul B. Bhasin and Joseph D. Warner, NASA Lewis Research Center. Chris Chorey, Sverdrup Technology,

Inc., Lewis Research Center Group, 2001 Aerospace Parkway, Brook Park, Ohio 44142.

16. Abstract

There is great interest in the application of thin film high temperature superconductors in high frequency

electronic circuits. A ring resonator provides a good test vehicle for assessing the microwave losses in the

superconductor and for comparing films made by different techniques. Ring resonators made of YBaaCU3OT.x

have been investigated on LaAIO 3 substrates. The superconducting thin films were deposited by sequential

electron beam evaporation of Cu, Y, and BaF 2 with a post anneal. Patterning of the superconducting film was

done using negative photolithography. A ring resonator was also fabricated from a thin gold film as a control.

Both resonators had a gold ground plane on the backside of the substrate. The ring resonators' reflection

coefficients were measured as a function of frequency from 33 to 37 GHz at temperatures ranging from 20 Kto 68 K. The resonator exhibited two resonances which were at 34.5 and 35.7 GHz at 68 K. The resonant

frequencies increased with decreasing temperature. The magnitude of the reflection coefficients is used in the

calculation of the unloaded Q-values. The performance of the evaporated and gold resonator are compared with

the performance of a laser ablated YBa2Cu3OT_x resonator. The causes of the double resonance are discussed.

17. Key Words (Suggested by Author(s))

Superconducting thin films

Sequential evaporation

Microwave passive circuits

18. Distribution Statement

Unclassified - Unlimited

Subject Category 33

19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of pages

Unclassified Unclassified 12

NASAFORM1626OCT86 *For sale by the National Technical Information Service, Springfield, Virginia 22161

22. Price*

A03